Radio frequency interference excision in solar dynamic spectra using variance-based spectral statistics

In this paper, a radio frequency interference (RFI) excision algorithm using variance-based analysis on power spectral statistics is proposed and tested on the solar data collected from our frequency agile solar radio-telescope (FASR) subsystem test-bed (FST). A threshold approach working on a proposed test statistic tau(x; f ), shown to be F-distributed, is developed to effectively identify the presence of non-Gaussian RFI in the Gaussian solar flare background. Detailed discussions on signal duty cycle and threshold setting are provided. Real-data experimental results are presented to demonstrate the robustness and effectiveness of the proposed algorithm.